Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[dudeabides]

I really feel for AdamH on this one.  The kingdom is:
Vineyard, Fool's Gold, Swindler, Urchin, Alchemist, Spy, Mandarin, Upgrade, Hoard, Nobles

Clearly, the strategy here is to convert an Urchin (or two) to Mercenary(ies) to thin your deck. 

I open Swindler/Urchin to Adam's Urchin/Urchin.

T3: I have 3 Coppers and 2 Estates, buy 2nd Urchin.  I draw Urchin/Swindler.  Adam plays one Urchin, does not connect, buys 3rd Urchin.
T4: I play Urchin/Swindler, get a Merc.  Swindler reveals Urchin, and I turn it into a Silver; and buy Potion.  Adam has 3 Coppers/2 Estates, buys 4th Urchin (but only has 3 thanks to my Swindling).
T5: I buy Alchemist, Adam doesn't connect, buys Swindler.
T6: I play Swindler, hit Urchin again, give Adam a Silver.  Adam plays Urchin, draws Silver (which would have been an Urchin, were it not for my Swindler), and concedes.

Here's the log:
http://www.gokosalvager.com/static/logprettifier.html?/20150118/log.5085f5130cf270038ff92212.1421597116880.txt

[DG]

Swindling mercenaries to curses is worse.

[TrojH]

I open Swindler/Urchin to Adam's Urchin/Urchin.

Thus proving once again, you make your own shuffle luck.

[AdamH]

Didn't see this before I made this thread, which includes this game as one of many I wanted to look closely at. Yeah there are serious questions I want answers to, but I don't know which thread they'll end up in, so I'll tag this here.

My resignation was purely emotional, I should have played the game out from a strictly competitive standpoint.

I mentioned this in the other thread, but I don't think Swindler is a good opening here. The odds of something like this are catastrophically low, arguably lower than the odds of not getting a Mercenary because your Swindler wasn't an Urchin. I can't back that up because I don't have the numbers, but now I'm curious as to what they are. Hmm...

Swindling mercenaries to curses is worse.

I mean, I know +1s speak louder than words, but I disagree with this. If I'm going to combat snark with snark then I'll just say that if you had a Mercenary to begin with then you're already better off than what happened here; but also, the only time where that actually hurts a player who is playing correctly and getting multiple Mercenaries (because he saw his opponeng buy Swindler) is when he hasn't really gotten thin yet, AND has only managed to get one Mercenary so far. In this case I'd argue that it's roughly the same as this, only you have less economy I guess? I guess that doesn't matter but ya know.



Oh well, I'm not looking for pity here. One does make one's own shuffle luck, it's true. Dudeabides was playing well so it's not undeserved.

[pacovf]

I mentioned this in the other thread, but I don't think Swindler is a good opening here. The odds of something like this are catastrophically low, arguably lower than the odds of not getting a Mercenary because your Swindler wasn't an Urchin. I can't back that up because I don't have the numbers, but now I'm curious as to what they are. Hmm...

Swindler in T3 or T4 has 1/6 chances to hit an urchin, assuming double urchin opening for the other player.
Urchin in T3 or T4 has 1/11 chance to draw an urchin, assuming you opened double urchin. Any other possibility, and it makes no difference whether that urchin was a swindler instead.

It gets trickier if you take T5 into account.

EDIT: let me specify that second sentence, because I fail at English. The odds of a T3 or T4 swindler failing to draw the urchin it would have drawn had the swindler been an urchin instead are 1/11. Which wasn't what I wrote, but might have been what you understood anyway, bless your soul.

[AdamH]

OK so these are interesting numbers, but I've thought about this a bit and let me see about some stuff.

If Swindler hits and Urchin, I feel like while that's a bad thing to have happen to you, it doesn't necessarily mean you aren't going to collide Urchins when you otherwise would have. For example, in that game, I was already screwed on T4, whether or not my opponent was going to play a Swindler on me.

In fact, if we're only going to look at T3 or T4 collisions here, if I'm the second player, the only way my Swindler could possibly matter is if I:

1. draw and play it on T3
2. hit my opponent's 11th card, which is one of his opening Attack-buys
3. their other Urchin is in their T4 hand

I think these number account for (2) and don't consider (1) or (3). If you're first player, then you get these odds given you draw Swindler on T4 but you also get to add them to the odds that you draw it on T3 and you prevent a collision that otherwise would have happened (which, granted, has more ways that it's possible).

In any case, I'm terrible at computing these things, but that makes it seem that the odds of you actually messing up your opponent's T3 or T4 Urchin collision are roughly the same as messing up your own T3 or T4 collision because your Swindler should have been an Urchin, which means your decision should probably be based on the rest of the kingdom? Maybe?

Or is there something wrong with my reasoning?

[Awaclus]

I think the early economy advantage you get from Swindler is worth the slightly smaller odds of getting a Mercenary. I also usually open Militia or Cutpurse over the second Urchin for the same reason.

[dondon151]

So does anyone seem to remember that turning Coppers into Curses is a reason to open Swindler, or...

All this talk about small chances to swindle opposing Urchins or missing Urchin/Attack collisions gives me the impression that we're missing the point. Doesn't double Urchin miss more than half the time anyway?

[AdamH]

Well it seems to me that if Swindler turns a Copper into a Curse, it doesn't matter all that much because it's just going to end up in the trash anyways. And even if it hits a Copper every time, at least there are still some treasures in your deck for economy, whereas once you have a Swindler, if you have no Village you'll always prefer to play Mercenary instead.

Obviously this depends on the board, but it makes me question people who say "Urchin/Swindler is better than Double Urchin so I always open like that"

[Awaclus]

Well it seems to me that if Swindler turns a Copper into a Curse, it doesn't matter all that much because it's just going to end up in the trash anyways. And even if it hits a Copper every time, at least there are still some treasures in your deck for economy, whereas once you have a Swindler, if you have no Village you'll always prefer to play Mercenary instead.

Obviously this depends on the board, but it makes me question people who say "Urchin/Swindler is better than Double Urchin so I always open like that"

It's going to end up in the trash, but it's going to take a while before it ends up in the trash. Plus the Swindler itself gives you $2.

[WanderingWinder]

I'm pretty sure Urchin/Swindler is almost always better than Urchin/Urchin, and I'm coming with reasons.

Urchin/Urchin has 3 things going for it: it cycles you faster, it gives you a 1/11 chance of hitting when you otherwise would not have (well, 1/11 if you play that urchin which would have been a swindler), and if they miss each other, you usually get to knock them to 4 cards from 5 twice.

If you have Swindler instead, there's approximately a 1/6 chance you disrupt their ability to get to Mercenary, by a LOT, by hitting one of their attacks. To be fair, when you miss, you are cycling them and making it slightly more likely they collide. Overall, I think this makes Swindler slightly better than Urchin... except that Swindler also makes $2, which is usually going to be better than drawing a card. On those rare boards where drawing a card on turn 3-4 is much better than $2, double Urchin might be right. But most of the time $2 is better, and I think even if they're equal, Swindler would be a little bit better.



Swindler hitting Mercenary before they can play it, or before they play it much is really a thing, too. But a minor effect.

[pacovf]

OK so these are interesting numbers, but I've thought about this a bit and let me see about some stuff.

If Swindler hits and Urchin, I feel like while that's a bad thing to have happen to you, it doesn't necessarily mean you aren't going to collide Urchins when you otherwise would have. For example, in that game, I was already screwed on T4, whether or not my opponent was going to play a Swindler on me.

In fact, if we're only going to look at T3 or T4 collisions here, if I'm the second player, the only way my Swindler could possibly matter is if I:

1. draw and play it on T3
2. hit my opponent's 11th card, which is one of his opening Attack-buys
3. their other Urchin is in their T4 hand

I think these number account for (2) and don't consider (1) or (3). If you're first player, then you get these odds given you draw Swindler on T4 but you also get to add them to the odds that you draw it on T3 and you prevent a collision that otherwise would have happened (which, granted, has more ways that it's possible).

In any case, I'm terrible at computing these things, but that makes it seem that the odds of you actually messing up your opponent's T3 or T4 Urchin collision are roughly the same as messing up your own T3 or T4 collision because your Swindler should have been an Urchin, which means your decision should probably be based on the rest of the kingdom? Maybe?

Or is there something wrong with my reasoning?

I am not making any claims about optimal strategy, I am only giving numbers.

You are right that I was assuming that the swindling player was player 1 (divide by 2 to get the odds for player 2; attacks usually reinforce first player advantage...). I was also assuming that you draw the swindler in T3 or T4 because T5 is a mess to predict; multiply by 5/6 to get a lower bound estimation, I guess.


As player 1, the odds of messing up a collision of urchins (instead of just hitting an urchin) are indeed quite low: 5/66 (~1/13, i.e., slightly below the odds of messing up your own collision), assuming that you draw swindler T3 or T4.
Of course, these odds are the same as the odds of swindler creating a collision that wouldn't have happened normally, so it's a wash. P2 is still looking at 5/11 chances of colliding their urchins.

Note though, that simply hitting an urchin with swindler (comparatively high odds, 1/6), even if it wasn't going to collide, can still delay your first merc by a shuffle. So that seems to be the more useful number to consider, IMHO.

Then there's the whole thing about swindler actually doing something.

tl;dr: you should probably ignore any effect that buying a different attack other than a second urchin has on the odds of getting a merc in T3 or T4.

[AdamH]

OK, this is starting to make some sense. Let's say that the odds of Swindler preventing a T3/T4 Mercenary are close enough to the odds that your Swindler should have been an Urchin to collide then that it's a wash. The odds of colliding anyways are less than 50% so Swindler's effect on the other 50% of cases should be considered.

And it's pretty clear that Swindler's attack effect has a larger effect when you consider things past T3/T4, since, you know, you actually draw the cards that have been Swindled. This is clear to me if you're the first player, and it's also clear that the math gets really messy if you're second player.

Either way, this gives me the tools to understand why people are saying Urchin/Swindler is a better opening, so I feel like I can make a more informed judgment about these situations when I want to open Urchin/Attack vs. opening double Urchin.

This has been educational. +1s for people who helped me understand this and who had the patience to deal with me.